Medical economic impact of tracheotomy patients on a hospital system

Abstract

Background:

Tracheotomy patients are a small portion of hospitalizations, but account for disproportionately high risk and costs. There are many complex decisions that go into the care of these patients, and practice variation is expected to be compounded in a health system. This study sought to characterize the medical economic impact of tracheotomy patients on the hospital system.

Methods:

A retrospective review of the health system’s hospital billing software was performed for 2013, and pertinent outcomes measures were tabulated.

Results:

There were 829 tracheotomies performed in the health system of seven hospitals, with total costs of $128,883,865. Average length of stay was 36.74 days for principal procedures, and 43.36 days for tracheotomy as secondary procedures. Mortality was ∼18% overall, and re-admissions were 10.93% for primary, and 14.36% for secondary procedures. A fairly wide variation in each category among the different hospitals was observed.

Conclusions:

There are potentially many factors that impact variations of care and outcomes in patients with tracheotomy. Due to their large economic impact and risks for morbidity and mortality, a formalized care pathway is warranted. Goals of the pathway should include understanding medical decisions surrounding these complex patients, monitoring pertinent outcomes, reducing practice variation, and improving the efficiency of compassionate care.

Keywords::

• Tracheotomy
• Tracheostomy
• Length of stay
• Readmission
• Mortality
• Economic impact
• Care pathway
• Population management
• Hospital
• Health system
• Outcomes
• Medical decision-making

Introduction

Hospital patients requiring tracheotomy have among the most severe diseases, account for a disproportionately long length of stay (LOS), have a high risk of mortality (ROM), and consume significant resources. These issues are recognized by the Centers for Medicare & Medicaid Services with specific diagnosis-related groups (DRG) that warrants higher payments, but these often do not cover the true costs of such patients. There are also many other complexities in the cycle of care for tracheotomy patients, and these have a potentially far greater impact on a hospital system where there can be practice variations among the many different healthcare providers at different hospitals.

In a prior study, the effect of tracheotomy timing on LOS at The Mount Sinai Hospital in New York was explored just prior to the acquisition of a large hospital system by our academic medical center1. This involved a retrospective review of 592 tracheotomies performed by an otolaryngology service from 2005–2010, 128 received early tracheotomy (by day 7 of ventilation with an average of day 4), and 464 late tracheotomy (by day 8 or beyond with an average of day 14). Both groups had comparable comorbidities, severity of illness (SOM) and ROM on admission. In the early tracheotomy group, duration of mechanical ventilation was 45% less (mean ± standard error: 21.47 ± 1.86 days vs 39.33 ± 1.33 days; p < 0.001), total intensive care unit (ICU) stay was shortened by 33% (17.52 ± 1.38 days vs 26.27 ± 0.73 days; p < 0.001), and length of overall hospital course was reduced by 34% (35.85 ± 2.57 days vs 54.28 ± 1.60 days; p < 0.001). There was a similarly impressive difference in the financial impact of these two groups on the hospital costs and reimbursement. With an average ICU patient requiring mechanical ventilation costs upwards of $4000 per day, the average reduction of 8 days in the ICU in the early compared to the late tracheotomy group in this study amounted to savings of $32,000 per patient.

There are many challenges when expanding from an academic hospital into a multi-hospital system to include different institutional cultures, practice patterns, patient populations, and concepts of best practice. We sought to investigate the medical economic impact of tracheotomy patients upon our new hospital system, and to begin to understand possible variations in care and patient populations.

Methods

A retrospective review was conducted through our hospital billing software to identify adult patients (aged 18 and older) undergoing tracheotomy in our hospital system over the calendar year 2013. This study is IRB exempt. International Classification of Disease (ICD-9) codes for tracheotomy 31.1, 31.19, and 31.74 were included, and stratified to primary procedure or secondary procedure (in-patients who had undergone other surgeries during their hospitalization). Demographic data was also collected. Information was stratified by individual hospital, total cases, total costs, cost/case, LOS, readmissions, and mortality. Cost data was determined through our hospital cost accounting system, Transition Systems Incorporated (TSI). The Premier Quality Advisor system (QA) was utilized to extract case volume, length of stay, mortality, readmission, and cost data. Mortality is defined as death during hospitalization, and for these patients the day of death is also the last day of LOS. The QA system receives monthly demographic, utilization, and cost data from hospitals 3 and 6; however hospitals 1, 2, 3, and 5 do not send cost data, only charges to QA, therefore costs presented represent a ratio of cost/charge. Costs presented include all fixed and variable direct and indirect costs allocated to each discharge. Variable costs are expenses that relate directly to or vary with the activity (volume) of the department. Fixed costs are expenses that do not relate directly to or vary with the activity (volume) of the department. Variable costs primarily include direct hands-on care, pharmaceuticals, imaging and laboratory testing, blood products, and medical–surgical supplies. Fixed costs are generally expenses related to overhead departments such as Information Technology, Human Resources and Finance, utility costs, and capital depreciation. One hospital in the system is a specialty eye and ear hospital and was not included in the analysis, as tracheotomy was not commonly performed at that site.

Results

There were a total of 829 tracheotomies performed during the study period of 2013. As demonstrated in Table 1, 383 of these were considered the principal procedure at the time of discharge, and 446 were secondary procedures. Total cost associated with these patients was $128,883,865 for an average cost/case of $155,469. Three-hundred and ninety-seven cases (47.89%) were performed at the flagship academic institution.

Table 1. Total Mount Sinai Health System (MSHS) tracheotomy cases in 2013, stratified by hospital.

Facility	Total cases	Total costs	Principal procedure	Secondary procedure	Total cases
MSHS	829	$128,883,865	383	446	100%
1	76	$6,120,982	41	35	9.17%
2	192	$17,681,489	135	57	23.16%
3	397	$78,847,669	106	291	47.89%
4	58	$9,692,850	33	25	7.00%
5	66	$12,216,362	50	16	7.96%
6	40	$4,324,514	18	22	4.83%

Tables 2 and 3 show the results for the six hospitals in the system, stratified by primary and secondary procedures, and listing cases, costs, LOS, readmissions, and mortality. The average cost/case was notably greater in the post-surgical patient at $168,017 compared to those undergoing tracheotomy as a primary procedure at $126,207. Similarly average LOS was 43.36 days in the post-surgical patient compared to 36.74 days in those undergoing the primary procedure. Wide variation was noted in cost/case at the different hospitals, LOS, and readmissions. Observed mortality also had some variation, with an overall average mortality among tracheotomy patients at 18%.

Table 2. Tracheotomy as principal procedure in 2013.

Facility	Cases	Total costs	Cost/case	LOS (days)	Re-admissions	Mortality
MSHS	383	$49,434,216	$126,207	36.74	10.93%	18.21%
1	41	$2,971,896	$72,485	41.71	18.52%	34.15%
2	135	$13,130,874	$92,724	21.11	9.68%	8.15%
3	106	$17,703,195	$165,242	45.69	8.43%	21.70%
4	33	$5,434,083	$164,669	42.47	3.57%	9.68%
5	50	$8,440,862	$165,808	50.70	14.29%	30.61%
6	18	$1,753,306	$97,406	40.94	28.57%	17.65%

Table 3. Tracheotomy as secondary procedure in 2013.

Facility	Cases	Total costs	Cost/case	LOS (days)	Readmissions	Mortality
MSHS	446	$79,449,649	$168,017.00	43.36	14.36%	17.9%
1	35	$3,149,086	$90,126.00	48.69	28.00%	28.57%
2	57	$4,550,615	$79,497.00	15.11	7.41%	5.26%
3	291	$61,144,474	$196,540.00	46.41	15.45%	18.82%
4	25	$4,258,767	$169,425.00	48.78	5.26%	24.00%
5	16	$3,775,500	$232,975.00	66.57	15.38%	13.33%
6	22	$2,571,208	$116,358.00	50.35	11.11%	18.18%

Discussion

Tracheotomy patients account for a small number of hospital admissions, but a disproportionately large portion of costs. In our hospital system, only 829 of these patients were responsible for almost $129 million in costs, almost $156,000 per case (not including professional services), and more than 18% mortality. Our results also demonstrate a fairly wide variation in LOS, cost/case, re-admission, and mortality between hospitals. These variations have yet to be explored, but may be explained by any number of factors, including differing patient demographics and comorbidities at the different hospitals, as well as different qualifications, philosophies, and standards of the treating physicians.

The concept of early vs late tracheotomy is apparently an important determinant of outcomes, and this has been explored in a number of studies. Park et al.2 studied factors determining the timing of tracheotomy, and found that Acute Physiology And Chronic Health Evaluation (APACHE) II scores ≥20 and time delay for tracheostomy were significantly associated with tracheotomy after 2 weeks. The finding is almost paradoxical implying that sicker patients who are more likely to require tracheotomy tended to receive the late procedure. Young et al.3 attempted a multi-center prospective study on determining the effect of early vs late tracheotomy on the primary outcome measure of mortality, and found no significant difference in early vs late tracheotomy groups. They concluded that clinicians did not have the ability to predict which patients required extended ventilatory support. In a Cochrane review, Gomes Silva et al.4 pooled four studies in which a total of 673 patients were randomized to either early or late tracheostomy. They found a high risk of bias in the studies, and no strong evidence for real differences between early and late tracheostomy in the primary outcome of mortality.

While it is controversial to perform tracheotomy early in patients that may still be able to wean off the ventilator, other outcomes such as voice, swallowing, and airway rehabilitation benefit from earlier tracheotomy. Prolonged intubation results in laryngeal inflammation and ulceration, placing the patient at increased risk of laryngotracheal stenosis5. Sedation needed for endotracheal intubation with ventilation also mandates sedentary behavior, leading to the propensity for decubitus ulcers, muscle deconditioning with atrophy, and delayed recovery of cognition.

These issues have led to a number of attempts at creating algorithms for tracheotomy care, although mostly anecdotal in the absence of formalized clinical practice guidelines. Protocols exist regarding tracheotomy timing6, medical decision-making, and candidacy for tracheostomy based on a standardized approach to ventilator weaning7, tracheotomy care and suctioning8, utility of the percutaneous tracheotomy technique9^,10, management of tracheotomy and laryngectomy airway emergencies11, and nursing care and management of tracheotomy emergencies12.

The diversity of these algorithms pays tribute to the interdisciplinary care required of these patients. Complexities in the cycle of care for tracheotomy patients include:

1. Absence of formalized clinical practice guidelines for each of the disciplines contributing to the care of tracheotomy patients;

2. Often a lack of seamless co-ordination between these disciplines;

3. Some heterogeneity in the patient population on mechanical ventilation;

4. Timing of performing tracheotomy in a ventilator-assisted patient;

5. Lack of definitions for early and late tracheotomy;

6. Technique, location, and qualifications of the team performing the tracheotomy;

7. Consistency of nursing and respiratory therapy experience with these patients;

8. Infection control issues that are common to patients with prolonged ventilation;

9. Family education and patient disposition following tracheotomy;

10. Differing medical insurance coverage regarding rehabilitation facility placement;

11. Potentially large variation in determining whether the patient is a candidate for decanulation;

12. Unknown rate of long-term complications associated with different tracheotomy techniques as surviving patients are transferred to long-term care facilities that are often not affiliated with the initial hospital; and

13. Cultural differences between different hospitals in a health system.

A tracheotomy or respiratory failure cycle of care pathway would be an ideal approach to explore these many issues in such a hospital system and reduce practice variation. Care pathways (also called critical or clinical pathways) structure the implementation of best practice, ideally with the use of clinical practice guidelines (CPGs) to create meaningful algorithms or protocols13. These can represent the medical decision-making process for providers, and also determine workflow for healthcare providers. The difference between a pathway and an algorithm is that the pathway has measureable outcomes at each step, and the pathway can be modified upon assessing these outcomes.

Burns et al.14 instituted an evidence-based clinical pathway for mechanically-ventilated patients in five intensive care units (ICU’s) at their 550-bed academic hospital, including protocols for weaning and sedation use. They analyzed outcomes in 510 patients on the pathway, compared to 595 historical controls, and found statistically significant differences in days on mechanical ventilation, ICU LOS, hospital LOS, and mortality, with a realized savings of over $3,000,000 in the 1-year study period. The study demonstrated the potential of a dedicated pathway with administrative support to improve outcomes and timing. While this approach can be extended to other areas of tracheotomy patients such as the complexities outlined above, the paucity of formal CPG’s limits the starting point for creation of care algorithms.

The population of patients requiring tracheotomy has a clear medical economic impact on hospital and health systems. We have demonstrated apparent practice and outcomes variation, perhaps in part due to our heterogeneous patient population. The many complexities in the cycle of care for tracheotomy patients can be addressed in conjunction with evidence-based care pathways, and this should yield vital insight into their more efficient population management. Such a system-wide approach should also help streamline patient flow, understand and diminish practice variation, and improve the value of medical care.

Transparency

Declaration of funding

The authors did not receive any funding for this study, nor do they have any conflicts of interest or other financial disclosures to make.

Acknowledgments

The authors wish to acknowledge Jodi Cohen in the President’s office at The Mount Sinai Health System for her assistance in data retrieval and analysis.